metadata
license: cc-by-nc-sa-4.0
base_model: ElnaggarLab/ankh-base
tags:
- generated_from_trainer
model-index:
- name: TooT-PLM-P2S
results: []
TooT-PLM-P2S
This model is a fine-tuned version of ElnaggarLab/ankh-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1451
- Q3 Accuracy: 0.7122
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0003
- train_batch_size: 1
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 6
- gradient_accumulation_steps: 4
- total_train_batch_size: 24
- total_eval_batch_size: 48
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.2
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Q3 Accuracy |
|---|---|---|---|---|
| 0.2036 | 1.0 | 449 | 0.1943 | 0.5833 |
| 0.1686 | 2.0 | 899 | 0.1864 | 0.5688 |
| 0.1597 | 3.0 | 1349 | 0.1770 | 0.5774 |
| 0.159 | 4.0 | 1799 | 0.1740 | 0.6245 |
| 0.1503 | 5.0 | 2248 | 0.1731 | 0.6851 |
| 0.1479 | 6.0 | 2698 | 0.1670 | 0.5961 |
| 0.1447 | 7.0 | 3148 | 0.1617 | 0.5936 |
| 0.1395 | 8.0 | 3598 | 0.1550 | 0.6307 |
| 0.1298 | 9.0 | 4047 | 0.1481 | 0.5573 |
| 0.1187 | 9.98 | 4490 | 0.1451 | 0.7122 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0
- Datasets 2.14.5
- Tokenizers 0.14.1